From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Greaves Subject: How to recover from a drive failure. Date: Fri, 12 Nov 2004 14:05:58 +0000 Message-ID: <4194C346.9000503@dgreaves.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from s2.ukfsn.org ([217.158.120.143]:31630 "EHLO mail.ukfsn.org") by vger.kernel.org with ESMTP id S262531AbUKLOGA (ORCPT ); Fri, 12 Nov 2004 09:06:00 -0500 Received: from localhost (lucy.ukfsn.org [127.0.0.1]) by mail.ukfsn.org (Postfix) with ESMTP id 1A3A8E6D56 for ; Fri, 12 Nov 2004 14:04:29 +0000 (GMT) Received: from mail.ukfsn.org ([127.0.0.1]) by localhost (lucy.ukfsn.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 01825-15 for ; Fri, 12 Nov 2004 14:04:29 +0000 (GMT) Received: from oak.dgreaves.com (modem-3248.putangitangi.dialup.pol.co.uk [81.78.204.176]) by mail.ukfsn.org (Postfix) with ESMTP id 72465E6A89 for ; Fri, 12 Nov 2004 14:04:28 +0000 (GMT) Received: from ash.dgreaves.com ([10.0.0.106]) by oak.dgreaves.com with esmtp (Exim 4.20) id 1CSc7x-0005z3-Ld for linux-ide@vger.kernel.org; Fri, 12 Nov 2004 14:10:13 +0000 Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: IDE Linux Hi I have raid5 setup and a disk died a couple of weeks back. The RMA return arrived this morning and, as luck would(n't) have it, another raid disk died too. The disks are Maxtor 250Gb SATA running through a Promise TX4. It's a shame SMART doesn't work yet - but they are running cool (12cm fan pointed at them) and well powered. Anyways, now I have a dead raid device :( Since I have a good disk I was going to: * insert new drive as /dev/sdd1 * dd if=/dev/sda1 of=/dev/sdd1 * mdadm /dev/md0 --remove /dev/sda1 * physically swap /dev/sda and /dev/sdd so /dev/sdd * mdadm /dev/md0 --add /dev/sda1 Is this the best approach? Especially the dd? I think I saw a 'better way' but can't find it despite some concerted googling. Any suggestions gratefully recv'd. David PS yes I'm buying a hot spare. Yes I wish I'd done that before but you have to balance risk and I didn't expect 2 new Maxtor drives to die within a few months of purchase. 1million hours MTBF - yeah, right. Equally why should a two sets of bad blocks screw up my raid array. No, I don't have a backup of 1Tb of data - I'm a home user.